Applied, thanks! On Wed, 9 Nov 2011, Alexandre Oliva wrote: > On Nov 7, 2011, Sage Weil <sage@xxxxxxxxxxxx> wrote: > > > New kernels (.39+) and new glibc have syncfs(2), which syncs only the > > fs the ceph-osd is serving up. > > I haven't seen the current glibc implementation, but the one I have on > my system is a stub that sets errno and returns nonzero. autoconf > correctly detects it as a stub, but I'm concerned that syncfs might fail > with say ENOSYS and we'd happily go ahead assuming the data is properly > sync()ed. > > The patch below fixes this, arranging for us to fallback to sync if > syncfs fails, but to try btrfs sync in between, under the assumption > that the ioctl will fail if fd isn't a btrfs inode. > > > http://linux.die.net/man/2/syncfs > > The man page is wrong or outdated: syncfs returns int, not void. > > -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
